1
$\begingroup$

So I do have data like this:

a

With the help of distinct order IDs, I can figure out how many orders are there and from units shipped, I can get the number of items in the order. Now I want to predict future orders having x items for different ranges (e.g. 1-10 items, 10-20 items, 20+ items). Like tomorrow there might be 6 orders one of them will have 1-10 items, 4 will have 10-20 items and one will have 20+ items. The problem is I can predict them (using LSTM for now) separately using data for only those kind of orders but they obviously do not sum up to the amount predicted by model for totals. So how can I link them up so that total sums up to the values from the individual models?

Approach I am currently using: I just predict orders for each item range and then manually sum them up. This way they do sum up but error accumulates

$\endgroup$

1 Answer 1

0
$\begingroup$

You should munge the data until you have a target value that aligns with the goal of the project. Then fit a model.

It sounds like you want the target to be a binned number of items {(1-10), (10-20), (20+)}. Each row should be unique Order_ID with one of those three target values.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.